Interaction of prokaryotes with their viruses (phages) and plasmids accounts for horizontal gene transfer (HGT) that underlies the spread of antibiotic resistance and emergence of human pathogens. Bacteria evolved numerous systems to limit HGT. A novel prokaryotic defense system against foreign DNA is based on CRISPR (clustered regularly interspaced short palindromic repeats) cassettes and cas genes. A CRISPR cassette consists of direct repeats interspersed with spacers of highly variable sequence. Small CRISPR RNAs (crRNAs) bound to a large Cas proteins complex recognize foreign DNA, matching the spacer sequence present in crRNA, and destroy it. This process is referred to as CRISPR interference. Spacers in CRISPR cassettes are excluded from interference. Viral or plasmid-derived DNA is acquired by CRISPR cassette, becoming a spacer, in a process called CRISPR adaptation. Acquisition of host-derived spacers must be avoided, for it will lead to self-interference. Neither stage of CRISPR response is fully understood. We propose to study CRISPR function in Escherichia coli, the best-studied prokaryote. CRISPR/cas loci of laboratory E. coli are dormant. We developed genetic systems to study both stages of E. coli CRISPR response. We will use these systems and genetic, biochemical, crosslinking, laboratory evolution, and modeling approaches to: Aim 1. Analyze CRISPR interference and identify rules that govern self versus non-self DNA recognition by CRISPR interference machinery; characterize in vitro Cas protein-crRNA complexes formed with foreign DNA targeted for degradation, and localize the sites of crRNA-mediated target cleavage. Experiments will be performed with existing systems targeting the M13 phage and with new systems interfering with lytic T-odd phages and RNA phages of E. coli. Aim 2. Analyze CRISPR adaptation and determine i) rules that govern self versus non-self DNA discrimination by CRISPR adaptation machinery; ii) sequences outside CRISPR cassette that affect spacer acquisition; and iii) molecular details of the process that leads to appearance of extra spacer-repeat units in CRISPR cassette. Cas protein complexes formed with foreign DNA targeted for adaptation will be characterized in vitro and in vivo by trapping them at protein roadblocks. To better understand CRISPR-mediated viral-host dynamics and co-evolution we will monitor spacer acquisition in CRISPR cassettes of the host and viral mutations that render CRISPR interference ineffective in continuously infected cultures and develop a mathematical model of this process in collaboration with a group of bioinformaticians. As a result of proposed work novel molecular mechanisms operational during CRISPR response will be revealed and new ways for strain engineering and gene silencing in prokaryotes will be developed. The significance of proposed work will not be limited to E. coli, since CRISPR loci are found in more than 40% eubacteria and in 95% of archaea.